W H I T E PA P E R

Cloud Computing:
Why Managing Performance Matters

www.gomez.com
Executive Summary:
Cloud computing is driving a fundamental shift in the way organizations build, deploy and
use applications, and it’s raising expectations on how quickly and cost-effectively new IT
functionality can be made available to the business. And even though the delivery chain
for these “borderless applications” now crosses organizational and geographic boundaries,
users will still expect the applications to perform well, and they will hold IT accountable
if they don’t. For its part, IT is faced with managing an increasingly complex and diverse
delivery chain, consisting of perhaps dozens of service and content providers spread
around the world
The challenge for IT is to meet the business expectation of faster delivery of new
functionality, while at the same time maintaining end-to-end visibility and control of
application performance and availability across a rapidly growing network of third-party
service providers.
Most current Application Performance Management (APM) solutions provide a narrow,
compartmentalized view, showing only individual components of the delivery chain,
with little or no integration between the components. However, customers do not want
different ways to manage application performance levels from different providers (or in
some cases, no way)  —  they require a single end-to-end view across the entire Web
application delivery chain, regardless of its physical, virtual or cloud attributes.
The only way to effectively meet these customer requirements is to provide a single
integrated solution that tests, monitors and manages application performance from
the only perspective which really matters: the end user’s.

Cloud computing 101:

Cloud computing provides on-demand, real-time network access to shared resources
that can be physically located anywhere across the globe. From an enterprise point of
view, it represents a fundamental shift in the way to acquire and implement new services
(computing power, storage, software, etc.) to support the business. Instead of internally
developed, monolithic systems, or lengthy and costly implementations of customized third-
party business solutions, cloud computing provides an agile and flexible environment with
shorter solution implementation cycles and much lower initial cost.
(A detailed description of cloud computing has been produced by the US National Institute of Standards and Technology
(NIST)  —  see Appendix A.)

The business benefits of cloud computing include lower IT costs — new hardware and
software functionality is acquired on a metered-service basis rather than through capital
expenditure — and increased business agility—application enhancements to support
changing business needs no longer rely on internally developed or customized third-party
software. Also, since cloud applications are inherently Web- and Internet-based, enterprises
can quickly and easily extend new functionality to customers via their Web sites.
As an example, consider a common financial application such as mortgage loan pre-
approval. As a legacy application, the client-based software contains all the business logic
necessary to guide the customer service representative (CSR) through the pre-approval
process. Several steps may require manual intervention — accessing property tax records,
FEMA data and credit history — because the information does not exist inside the enterprise
and can change very quickly. The net result is either a delay in pre-approval, or an inaccurate
estimate of closing and servicing costs to the customer, both of which can result in lost
business. Provided that the legacy application uses a modern architecture (.NET for instance),
it is relatively straightforward to add various Web (cloud) services to the application to access
the missing information, resulting in faster (and more accurate) decision making.

W H I T E PA P E R — C l o u d C o m p u t i n g : W h y M a n a g i n g P e r f o r m a n c e M at t e r s 2

Most analysts agree that performance
and availability are two of the key
inhibitors to the widespread adoption
of cloud computing.
IDC conducted a survey of the IDC
Enterprise panel in Q3 2009.
This panel ranked security, availability
and performance as the top three
concerns for cloud adoption:
• Security – 87.5%
• Availability – 83.3%
• Performance – 82.9%
Moreover, 88.6% stated that cloud
service providers need to provide SLAs.
Existing enterprise applications that have been upgraded with cloud-based services are
commonly known as “composite applications.” They offer a “best of both worlds” approach
to the business: They protect the significant investment in existing software, but they also
provide a rapid, relatively inexpensive way to add functionality to meet changing
business needs.
Additionally, with a browser-based user
interface, the application can be offered
to customers as a self-service option
through the enterprise’s Web site — this
can have a significant positive impact on
the bottom line, since the site is open for
business 24x7. In the case of mortgage
pre-approvals, for instance, a number
of financial institutions now offer Web-
based applications which can provide
on-the-spot decisions, with a complete,
accurate listing of closing and servicing
costs. The application saves money
because no CSR is involved in the pre-approval process, and boosts business because
consumers are less likely to shop around if they receive a fast, accurate pre-approval.
Cloud computing actually provides even greater flexibility – and potential cost savings — by
allowing enterprises not only to leverage existing applications in the cloud (Software-as-a-
Service); it also allows the enterprise to deploy applications on a service provider’s servers
(Platform-as-a-Service), and to leverage the availability of processing, storage, networks and
other fundamental computing resources owned by service providers
(Infrastructure-as-a-Service).
Every silver lining has its dark cloud:
Cloud-based applications offer improved agility and
lower IT costs by extending the application delivery
chain beyond the enterprise firewall to leverage
complementary software and services from cloud
service providers. The trade-off is IT management
complexity. Previously, IT had complete control of
applications and data because
everything resided in the data center. The arrival of
these “borderless” applications
presents a completely new challenge:
How can IT manage the performance of applications and their availability
to the business when those applications are now dependent on services and
content that cross organizational and geographic borders?
Service providers are typically unwilling to commit to specific Service Level Agreements,
and for those that do, there is a lot of inconsistency — and confusion — in their definitions
of performance and availability. Amazon for instance currently quotes availability in terms
of “outages” — periods of five minutes or more during the service year in which Amazon
EC2 was in the state of “region unavailable.” Others prefer to quote more general statistics
1 Frank Gens, IDC Research “The Maturing Cloud—What it Will Take to Win,” March 4, 2010
W H I T E PA P E R — C l o u d C o m p u t i n g : W h y M a n a g i n g P e r f o r m a n c e M at t e r s 3

CloudSleuth Web Portal
The Compuware-sponsored CloudSleuth
community Web Portal is designed to
meet the growing need for authoritative,
objective measurements of cloud service
providers. It provides free access to
real-time performance and availability
visualizations of leading cloud providers
around the world, plus other valuable
data such as blogs, forums and white
papers—all focused on best practices
for building, deploying, and managing
cloud-based applications.
such as “multiple redundant gigabit Internet connections” and “greater than 99.95% service
availability.” To put these figures into context, a 99.95% availability means that unplanned
downtime of a cloud-based service will average no more than 12 minutes per month.
Compare this with about 95 minutes per month of downtime for the average exchange
server, and the initial reaction is that there’s no need to worry about performance and
availability. However, this is a very dangerous assumption, since it ignores a critically
important point: The service provider is just one part of the Web application delivery chain.
From an end-user perspective, poor performance or non-availability of an application looks
exactly the same, irrespective of where the problem actually is in the application delivery
chain — one of the service providers, the data center, the network, the enterprise, the
cloud or the end user’s own device — and has exactly the same productivity impact to the
business. In the earlier example of mortgage loan pre-approval application which utilizes
cloud-based services, what happens if one of the services performs badly or is not available
at all? How can the enterprise determine if it’s an internal data center, Internet, 3rd-party
vendor, or an end-user browser problem?
To further complicate things, geographic location can also have a dramatic impact on
the overall performance of a cloud-based application — this is somewhat contrary to the
popular belief that Internet communication is virtually instantaneous. A worse-case scenario
is that “all lights are green” in the data center, but some (not all) customers are complaining
about performance issues. Without detailed fault-domain information across the entire
delivery chain, it is virtually impossible to isolate and fix performance and availability issues
in a timely manner, before they start to impact users.
The Earth isn’t flat:
To illustrate how all the components of the delivery chain can impact performance of a
Web-based application, the graphs below show actual measurements from CloudSleuth,
the Compuware Gomez-sponsored Web Portal that provides real-time visualizations of
the performance and availability of leading cloud service providers. CloudSleuth measures
performance of a simple application (no I/O- or CPU-intensive tasks) deployed anonymously
at a number of leading cloud services providers.
“Last Mile” (ISP) performance:
This test shows how the response time is impacted
by the performance of the user’s ISP (the so-called
“last mile” connection).
Note that users in Wyoming are experiencing
performance issues because of “last mile” connectivity
problems, not because of Amazon.
Geography:
The graphs clearly show that the farther away the
user is from the application, the longer the response
time.
This test also illustrates that if enterprises have a
choice of service providers, it is best to choose one
that is nearest to their user and/or customer base.
W H I T E PA P E R — C l o u d C o m p u t i n g : W h y M a n a g i n g P e r f o r m a n c e M at t e r s 4
 Time of Day:
This test shows that the performance of cloud service providers is not constant, but can vary quite widely
throughout the day. This is generally because the service provider is handling a varying load from other users
on their systems.

It also illustrates another practical point about cloud performance: The cloud theoretically provides “rapid
elasticity,” meaning that wide variations in load can be accommodated without significantly impacting the
performance of individual applications. In reality, though, cloud service providers have to live by the same
rules of economics as everyone else—they do not have banks of servers lying idle to cope with these peaks in
demand. Although applications operate in their own “instances” at the service provider, their performance is
affected by what their neighbors are doing!
Impact of Time of Day on Preformance

The Gomez Performance Network is then used to access those applications from backbone
and “last mile” locations around the world to provide actual performance results. All the
tests below use only Amazon EC2 East and West.

Putting it all together:

Fred Smith, the founder of FedEx, once remarked: “Information about the package is as
important as the package itself.” He was making the case that it’s not enough to provide
a general statement of service quality; you must be able to present information on the
particular service you are delivering to a particular customer at a particular time, irrespective
of where the package is. The same is true for “borderless” applications — these require a
solution that can monitor and manage application performance irrespective of physical,
virtual or cloud attributes. Traditional enterprise application performance management tools
are unsuited to the task of managing this new generation of applications, because they
only provide narrow, technology-centric keyhole views into the performance of specific
components or processes. The only way to truly solve performance and availability problems
is through a holistic view of application performance that encompasses the entire Web
application delivery chain.

Compuware Gomez addresses the challenges of

cloud computing:
Gomez provides the first Web Performance Management solution capable of managing
performance and availability across the entire cloud service delivery chain. Gomez provides
a unique, end-to-end view of the entire Web application delivery chain from within a
single unified dashboard. IT administrators can use one “pane of glass” to see performance
bottlenecks wherever they arise. For best practices and details on ways to measure cloud
application performance end user experience, please see the Gomez White Paper: “Cloud
Computing and Web Application Performance: The Risks of Adopting The Cloud.”
The Gomez Platform can help you evaluate cloud providers, test your cloud application prior
to deployment, and monitor it after it goes into production.

Using Gomez can help you:

• Monitor performance and availability of cloud-deployed applications Evaluate and
select cloud vendors and technologies based on performance and reliability
• Measure impact of cloud infrastructure and configuration on end-user experience
under normal and peak conditions
• Validate cloud-bursting configurations and solutions
• Monitor SLA compliance
• Optimize performance in cloud application delivery chain while reducing cost

W H I T E PA P E R — C l o u d C o m p u t i n g : W h y M a n a g i n g P e r f o r m a n c e M at t e r s 5
Summary:
Companies are anxious to evaluate and realize the business benefits of cloud computing but
are also mindful of the potential business risks that the technology exposes. Gomez’s Web
Performance Management solutions provide an innovative approach to the Web application
performance and availability challenges inherent in cloud computing. It offers the three key
elements essential to meet these challenges: Visibility across the entire delivery chain; deep-
dive troubleshooting for isolation and resolution; and unified dashboards for a single source
of truth relevant to different stakeholders.
Using Compuware’s Gomez solution, customers are now able to move ahead with
evaluating all aspects of cloud computing without being faced with unacceptable trade-offs
on service levels and security. No other vendor is able to offer the same features in a single,
integrated solution.

About Compuware Gomez

Gomez, the web performance division of Compuware, provides the industry’s leading
solution for optimizing the performance, availability, and quality of Web and mobile
applications. The on-demand Gomez platform integrates web load testing, Web
performance management, web cross-browser testing, and web performance business
analysis, enabling organizations to test from the “outside-in” – across all users, browsers,
devices, geographies, and data centers – using a global network of over 150,000 locations.
Gomez is an integral part of Compuware’s end-to-end Application Performance
Management offering, the industry’s only solution for optimizing application performance
across the Enterprise and the Internet. Over 3,000 customers worldwide, ranging from small
companies to large enterprises – including 12 of the top 20 most visited US websites – use
Gomez to increase revenue, build brand loyalty and decrease costs.

W H I T E PA P E R — C l o u d C o m p u t i n g : W h y M a n a g i n g P e r f o r m a n c e M at t e r s 6
Appendix A:
The NIST Definition of Cloud Computing
Authors: Peter Mell and Tim Grance
Version 15, 10-7-09
National Institute of Standards and Technology, Information Technology Laboratory

Note 1:
Cloud computing is still an evolving paradigm. Its definitions, use cases, underlying
technologies, issues, risks and benefits will be refined in a spirited debate by the
public and private sectors. These definitions, attributes and characteristics will evolve and
change over time.
Note 2:
The cloud computing industry represents a large ecosystem of many models, vendors, and
market niches. This definition attempts to encompass all of the various cloud approaches.

Definition of Cloud Computing:

Cloud computing is a model for enabling convenient, on-demand network access to
a shared pool of configurable computing resources (e.g., networks, servers, storage,
applications and services) that can be rapidly provisioned and released with minimal
management effort or service provider interaction. This cloud model promotes availability
and is composed of five essential characteristics, three service models and four
deployment models.

Essential Characteristics:
On-demand self-service. A consumer can unilaterally provision computing capabilities,
such as server time and network storage, as needed automatically without requiring human
interaction with each service’s provider.
Broad network access. Capabilities are available over the network and accessed through
standard mechanisms that promote use by heterogeneous thin or thick client platforms (e.g.,
mobile phones, laptops and PDAs).
Resource pooling. The provider’s computing resources are pooled to serve multiple
consumers using a multi-tenant model, with different physical and virtual resources
dynamically assigned and reassigned according to consumer demand. There is a sense of
location independence in that the customer generally has no control or knowledge over the
exact location of the provided resources but may be able to specify location at a higher level
of abstraction (e.g., country, state or data center). Examples of resources include storage,
processing, memory, network bandwidth and virtual machines.
Rapid elasticity. Capabilities can be rapidly and elastically provisioned, in some cases
automatically, to quickly scale out and rapidly released to quickly scale in.
To the consumer, the capabilities available for provisioning often appear to be
unlimited and can be purchased in any quantity at any time.
Measured Service. Cloud systems automatically control and optimize resource use by
leveraging a metering capability at some level of abstraction appropriate to the type of
service (e.g., storage, processing, bandwidth and active user accounts). Resource usage can
be monitored, controlled and reported providing transparency for both the provider and
consumer of the utilized service.

W H I T E PA P E R — C l o u d C o m p u t i n g : W h y M a n a g i n g P e r f o r m a n c e M at t e r s 7
Service Models:
Cloud Software as a Service (SaaS). The capability provided to the consumer is to
use the provider’s applications running on a cloud infrastructure. The applications are
accessible from various client devices through a thin client interface such as a web browser
(e.g., web-based e-mail). The consumer does not manage or control the underlying cloud
infrastructure including network, servers, operating systems, storage or even individual
application capabilities, with the possible exception of limited user-specific application
configuration settings.
Cloud Platform as a Service (PaaS). The capability provided to the consumer is to deploy
onto the cloud infrastructure consumer-created or acquired applications created using
programming languages and tools supported by the provider. The consumer does not
manage or control the underlying cloud infrastructure including network, servers, operating
systems or storage, but has control over the deployed applications and possibly application
hosting environment configurations.
Cloud Infrastructure as a Service (IaaS). The capability provided to the consumer is to
provision processing, storage, networks and other fundamental computing resources where
the consumer is able to deploy and run arbitrary software, which can include operating
systems and applications. The consumer does not manage or control the underlying cloud
infrastructure but has control over operating systems, storage, deployed applications and
possibly limited control of select networking components (e.g., host firewalls).

Deployment Models:
Private cloud. The cloud infrastructure is operated solely for an organization. It may be
managed by the organization or a third party and may exist on premise or off premise.
Community cloud. The cloud infrastructure is shared by several organizations and supports
a specific community that has shared concerns (e.g., mission, security requirements, policy
and compliance considerations). It may be managed by the organizations or a third party
and may exist on premise or off premise.
Public cloud. The cloud infrastructure is made available to the general public or a large
industry group and is owned by an organization selling cloud services.
Hybrid cloud. The cloud infrastructure is a composition of two or more clouds (private,
community or public) that remain unique entities but are bound together by standardized
or proprietary technology that enables data and application portability (e.g., cloud bursting
for load-balancing between clouds).
Note: Cloud software takes full advantage of the cloud paradigm by being service oriented with a focus on statelessness, low cou-
pling, modularity and semantic interoperability.

W H I T E PA P E R — C l o u d C o m p u t i n g : W h y M a n a g i n g P e r f o r m a n c e M at t e r s 8